sovant 1.1.0__tar.gz → 1.2.0__tar.gz

{sovant-1.1.0/src/sovant.egg-info → sovant-1.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sovant
-Version: 1.1.0
+Version: 1.2.0
 Summary: Sovant Memory-as-a-Service Python SDK
 Author: Sovant
 License: MIT
@@ -16,7 +16,8 @@ Dynamic: license-file
 
 # Sovant Python SDK
 
-Sovant is Memory-as-a-Service: a durable, queryable memory layer for AI apps with cross-model recall, hybrid (semantic + deterministic) retrieval, and simple SDKs.
+**Sovant is a governed AI memory layer for AI agents and applications.**
+Use it to store, search, and recall memories with profile awareness and enterprise-grade control over how memory is captured and used.
 
 [![PyPI version](https://img.shields.io/pypi/v/sovant.svg)](https://pypi.org/project/sovant/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
@@ -57,6 +58,32 @@ updated = client.memory.update(mem["id"], {
 client.memory.delete(mem["id"])
 ```
 
+## Recall vs Search
+
+Sovant provides two ways to query memories:
+
+- **`memory_recall()`** – Hybrid recall, profile-aware
+  Use for conversational queries: "What do you know about me?", "What happened on Project X?".
+  Uses Sovant's hybrid pipeline (profile fast-path + thread-scoped lexical + vector semantic search) and prioritizes profile facts (name/age/location) when available.
+
+- **`memory_search()`** – Semantic search
+  Use for topic lookup and discovery. Pure vector similarity search without profile logic.
+  Behavior unchanged from previous versions.
+
+```python
+# Recall for conversational queries
+context = client.memory_recall(
+    query="what do you know about me?",
+    limit=10
+)
+
+# Search for topic discovery
+topics = client.memory_search({
+    "query": "project updates",
+    "limit": 5
+})
+```
+
 ## Chat in 60 Seconds
 
 Stream real-time chat responses with memory context:

{sovant-1.1.0 → sovant-1.2.0}/README.md

@@ -1,6 +1,7 @@
 # Sovant Python SDK
 
-Sovant is Memory-as-a-Service: a durable, queryable memory layer for AI apps with cross-model recall, hybrid (semantic + deterministic) retrieval, and simple SDKs.
+**Sovant is a governed AI memory layer for AI agents and applications.**
+Use it to store, search, and recall memories with profile awareness and enterprise-grade control over how memory is captured and used.
 
 [![PyPI version](https://img.shields.io/pypi/v/sovant.svg)](https://pypi.org/project/sovant/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
@@ -41,6 +42,32 @@ updated = client.memory.update(mem["id"], {
 client.memory.delete(mem["id"])
 ```
 
+## Recall vs Search
+
+Sovant provides two ways to query memories:
+
+- **`memory_recall()`** – Hybrid recall, profile-aware
+  Use for conversational queries: "What do you know about me?", "What happened on Project X?".
+  Uses Sovant's hybrid pipeline (profile fast-path + thread-scoped lexical + vector semantic search) and prioritizes profile facts (name/age/location) when available.
+
+- **`memory_search()`** – Semantic search
+  Use for topic lookup and discovery. Pure vector similarity search without profile logic.
+  Behavior unchanged from previous versions.
+
+```python
+# Recall for conversational queries
+context = client.memory_recall(
+    query="what do you know about me?",
+    limit=10
+)
+
+# Search for topic discovery
+topics = client.memory_search({
+    "query": "project updates",
+    "limit": 5
+})
+```
+
 ## Chat in 60 Seconds
 
 Stream real-time chat responses with memory context:

{sovant-1.1.0 → sovant-1.2.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "sovant"
-version = "1.1.0"
+version = "1.2.0"
 description = "Sovant Memory-as-a-Service Python SDK"
 readme = "README.md"
 requires-python = ">=3.10"

{sovant-1.1.0 → sovant-1.2.0}/src/sovant/client.py

@@ -162,9 +162,27 @@ class Sovant:
             params['to_date'] = q.to_date
         return self._request("GET", f"{self.base_url}/api/v1/memory/search", params=params)
 
-    def memory_recall(self, q: SearchQuery):
-        """Semantic search alias for memory_search()"""
-        return self.memory_search(q)
+    def memory_recall(self, query: str, thread_id: str | None = None, limit: int | None = None):
+        """
+        Hybrid recall with profile awareness
+
+        Uses multi-stage pipeline (profile fast-path + lexical + semantic)
+        Guarantees profile facts (name/age/location) when available
+
+        Use recall() for conversational queries ("who am I?", "what do you know about me?")
+        Use memory_search() for pure semantic topic lookup
+
+        Args:
+            query: The search query (required)
+            thread_id: Optional thread context for thread-scoped recall
+            limit: Maximum results to return (default 8, max 50)
+        """
+        params = {'query': query}
+        if thread_id:
+            params['thread_id'] = thread_id
+        if limit:
+            params['limit'] = str(limit)
+        return self._request("GET", f"{self.base_url}/api/v1/memory/recall", params=params)
 
     def memory_update(self, id: str, patch: Dict[str, Any]):
         # Convert data field to content field if present

{sovant-1.1.0 → sovant-1.2.0/src/sovant.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sovant
-Version: 1.1.0
+Version: 1.2.0
 Summary: Sovant Memory-as-a-Service Python SDK
 Author: Sovant
 License: MIT
@@ -16,7 +16,8 @@ Dynamic: license-file
 
 # Sovant Python SDK
 
-Sovant is Memory-as-a-Service: a durable, queryable memory layer for AI apps with cross-model recall, hybrid (semantic + deterministic) retrieval, and simple SDKs.
+**Sovant is a governed AI memory layer for AI agents and applications.**
+Use it to store, search, and recall memories with profile awareness and enterprise-grade control over how memory is captured and used.
 
 [![PyPI version](https://img.shields.io/pypi/v/sovant.svg)](https://pypi.org/project/sovant/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
@@ -57,6 +58,32 @@ updated = client.memory.update(mem["id"], {
 client.memory.delete(mem["id"])
 ```
 
+## Recall vs Search
+
+Sovant provides two ways to query memories:
+
+- **`memory_recall()`** – Hybrid recall, profile-aware
+  Use for conversational queries: "What do you know about me?", "What happened on Project X?".
+  Uses Sovant's hybrid pipeline (profile fast-path + thread-scoped lexical + vector semantic search) and prioritizes profile facts (name/age/location) when available.
+
+- **`memory_search()`** – Semantic search
+  Use for topic lookup and discovery. Pure vector similarity search without profile logic.
+  Behavior unchanged from previous versions.
+
+```python
+# Recall for conversational queries
+context = client.memory_recall(
+    query="what do you know about me?",
+    limit=10
+)
+
+# Search for topic discovery
+topics = client.memory_search({
+    "query": "project updates",
+    "limit": 5
+})
+```
+
 ## Chat in 60 Seconds
 
 Stream real-time chat responses with memory context: